Full Length Research Paper

G×E interaction effects on yield and yield components of cassava (landraces and improved) genotypes in the savanna regions of Nigeria

O. O. Aina^1,2,3*, A. G. O. Dixon¹, Ilona Paul¹ and E. A. Akinrinde³

¹International Institute of Tropical Agriculture, P. M. B. 5320, Ibadan, Nigeria.

²College of Agricultural Sciences, Joseph Ayo Babalola University, Ikeji Arakeji, P. M. B. 5006, Osun State, Nigeria.

³Department of Agronomy, University of Ibadan, Nigeria.

*Corresponding author. E-mail: f_aina2002@yahoo.com.

Accepted 5 June, 2009

   Abstract

Genetic enhancement of cassava aimed at increasing production and productivity through the provision of broad-based improved germplasm is a major goal for cassava breeders. At the International Institute of Tropical Agriculture (IITA), Nigeria, 18 varieties comprising 12 landraces and 6 broad-based and improved varieties were evaluated at 4 locations in 3 years in a randomized complete block design in 4 replicates to determine variability among cultivars for yield components and adaptation to different environments. Results showed fresh root yield was significantly correlated (P < 0.001) with number of roots, harvest index, shoot weight and number of stands harvested. AMMI analysis partitioned main effects into genotypes, environments and G × E with all the components showing highly significant effects (P < 0.001). Environment had the greatest effect (70.3%), G × E interaction (19.0%) and genotype (10.7%). AMMI1 and unadjusted means selected the same winner in 9 out of 12 environments (75%), but differently in 3 environments. The GGE biplot (E and GxE interaction) delineated environments into mega-environments. Cultivar 4(2)1425 (moderately yielding) was the most stable and specifically adapted to Zaria. ABBEY-IFE, ATU-IWO and 2ND-AGRIC though moderately yielding were highly tolerant to CMD, suggesting a rich resource within the germplasm that could be enhanced for further genetic improvement of the crop.

Key words: Cassava, AMMI analysis, GGE biplot, GxE interaction, mega-environment.